NVDA 155.46 ▼0.44%GOOGL 280.20 ▼1.24%MSFT 506.24 ▼2.27%AMD 503.41 ▼0.17%INTC 55.17 ▼0.54%TSMC 359.69 ▼3.68%AMZN 257.89 ▲0.05%META 618.87 ▼1.20%AAPL 351.91 ▼0.25%PLTR 156.56 ▼2.39%
Markets at last close

Google · Security

Google warns of rising LLMjacking costs

·1 min read

Google Threat Intelligence Group has reported a rise in LLMjacking attacks, with fraudulent cloud charges exceeding USD $100,000 in a single day. Mandiant investigated cases in 2026 across North America, South America and Asia Pacific, affecting financial services, retail, technology and transportation companies. Some organisations incurred costs of as much as USD $1 million before access was cut off.

Attackers gained entry through exposed model endpoints, stolen cloud credentials, leaked API keys and phishing that captured active access tokens. They searched internal repositories and cloud systems for additional secrets, created unauthorised accounts and escalated privileges to maintain access. In several incidents, attackers enabled previously inactive model services before generating heavy inference traffic.

Investigators also identified an underground resale market involving illicit gateways called “LLM Mixers,” which aggregate and resell stolen model access. A service called Jet Night allegedly used stolen third-party API keys to make more than 500,000 API calls in less than eight hours, pushing close to USD $1 million in token costs onto a victim organisation.

Google recommends treating unauthorised activation of LLM services as a critical security event. Defensive measures include least-privilege access, secret scanning, stronger API key controls, rate limiting, budget alerts, phishing-resistant multi-factor authentication and gateway patching. Monitoring should also flag small validation requests that may precede large-scale abuse.

Originally reported by securitybrief.asiaRead the source →
Related coverage
All Google news →